Predictive equations using regression analysis of pulmonary function for healthy children in Northeast China.

Ya-Nan Ma,Miao-Miao Liu,Yu-Qin Liu,Ya-Dong Zhao,Guang-Hui Dong,Yang Zhao,Da Wang,Yungling Leo Lee,Jing Wang,Qin-Cheng He,Wan-Hui Ren,Robert K Hills

doi:10.1371/journal.pone.0063875

Abstract

BackgroundThere have been few published studies on spirometric reference values for healthy children in China. We hypothesize that there would have been changes in lung function that would not have been precisely predicted by the existing spirometric reference equations. The objective of the study was to develop more accurate predictive equations for spirometric reference values for children aged 9 to 15 years in Northeast China.Methodology/Principal FindingsSpirometric measurements were obtained from 3,922 children, including 1,974 boys and 1,948 girls, who were randomly selected from five cities of Liaoning province, Northeast China, using the ATS (American Thoracic Society) and ERS (European Respiratory Society) standards. The data was then randomly split into a training subset containing 2078 cases and a validation subset containing 1844 cases. Predictive equations used multiple linear regression techniques with three predictor variables: height, age and weight. Model goodness of fit was examined using the coefficient of determination or the R2 and adjusted R2. The predicted values were compared with those obtained from the existing spirometric reference equations. The results showed the prediction equations using linear regression analysis performed well for most spirometric parameters. Paired t-tests were used to compare the predicted values obtained from the developed and existing spirometric reference equations based on the validation subset. The t-test for males was not statistically significant (p>0.01). The predictive accuracy of the developed equations was higher than the existing equations and the predictive ability of the model was also validated.Conclusion/SignificanceWe developed prediction equations using linear regression analysis of spirometric parameters for children aged 9–15 years in Northeast China. These equations represent the first attempt at predicting lung function for Chinese children following the ATS/ERS Task Force 2005 guidelines on spirometry standardization.

Highlights

Lung function tests have become an indispensable tool for clinical evaluation of respiratory health and diseases [1]
Gender, age, weight and height are known to have a key impact on spirometric parameters [26]
Reliable interpretation of spirometric parameters depends on the availability of predictive equations for spirometric reference values in assessing the severity and nature of functional impairments [28]

Summary

Introduction

Lung function tests have become an indispensable tool for clinical evaluation of respiratory health and diseases [1]. Spirometry is a relatively simple, non-invasive method for measuring the ow and volume of air in the lung at maximal ination as a function of time using forced manoeuvres [2] It is widely accepted as a clinical tool for diagnosing obstructive, restrictive or mixed ventilatory defects such as chronic obstructive pulmonary diseases, interstitial lung diseases and asthma [3]. There may be changes in lung function that may not be precisely predicted by the existing reference equations [16], and there is an urgent need to develop equations that would more accurately predict spirometric reference values for Chinese children. The objective of the study was to develop more accurate predictive equations for spirometric reference values for children aged 9 to 15 years in Northeast China

Objectives

Methods

Results

Conclusion